Cosmic-rays (high-energy particles such as protons and electrons) form the link between the most extreme accelerators in our universe and the most fundamental processes in high-energy physics. Their complicated propagation through galactic magnetic fields lies at the heart of many significant questions in astrophysics. Investigations into the flux of cosmic-ray antimatter, the origin of high-energy neutrinos, and even the nature of dark matter, require detailed knowledge of cosmic-ray propagation. Observations by the Alpha Magnetic Spectrometer and the Fermi Large-Area Telescope have produced precise observations of GeV cosmic-rays. These allow us to constrain both the cosmic-ray flux at Earth, as well as the production of gamma rays by cosmic-ray interactions throughout the galaxy. Recently, new instruments -- including the High Altitude Water Cherenkov and High Energy Stereoscopic System have pushed our observations into the TeV range. To take advantage of this observational revolution, our models of cosmic-ray propagation must match the precision of current data. The parameter space of cosmic-ray diffusion is complex and multi-dimensional. Numerical algorithms have been developed to solve the time- and spatially-dependent diffusion equation, moving cosmic-rays from their natal sites (supernova remnants and pulsars) throughout the bulk of the galaxy. TeV datasets add new complexity -- while GeV cosmic-rays travel long distances before interacting (smoothing out local effects), TeV cosmic rays react strongly to local inhomogeneities. We propose a multifaceted effort to model cosmic-rays throughout the Milky Way. Our program has four objectives: (a) We will continue ongoing investigations into the effects of nuclear cross-section and solar-modulation uncertainties on low-energy cosmic-ray propagation. This study (currently running on SNIC) will set strong limits on dark matter annihilation and illuminate the role of solar dynamics in influencing the local cosmic-ray flux. (b) We will develop new multi-messenger models and data analysis of transient events, with the goal of correlated excesses across multiple experimental probes to develop more detailed models of high energy sources. (c) We will perform the first code comparison focused on the treatment of inhomogeneous diffusion by the two primary cosmic-ray algorithms employed in the community (Galprop and Dragon). This effort is unique to Stockholm – both PI-Linden and Co-I Korsmeier have significant expertise with Galprop, while Co-I de la Torre Luque is a member of the Dragon collaboration. (d) We will study stochastic electron cooling, developing new techniques that improve upon models which approximate cooling as continuous. Our group have recently shown that these models are inaccurate, and will develop new high-precision models for next generation data. Our efforts have a high probability of success. Our codes are developed, tested, and already run on SNIC resources. The remaining hurdle is the computational time required to marginalize our models over the parameter space of cosmic-ray propagation. We are confident that each objective will result in several published papers, with results produced within the one year NAISS allocation.